The interaction of the system components, especially the UCs, to the rest of the system that deliver DC power to the POD is critical for the proposed model and this interaction must be tested under controlled conditions. The objectives of this task are (1) development of power input variability curves based on historical wind data and turbine specifications, and (2) characterization of the system components including the OMU, PPS, EMU and the two PODs under controlled input power conditions, and (3) demonstration of ride-through capability by powering a POD directly with UCs .

AWS offers a historical simulation of power output from hypothetical wind farms. It was generated using AWS mesoscale numerical weather prediction models and simulates the energy generated from hypothetical (non-existent) wind farms in 10 minute intervals. The data spans 3 years 2004-2006 and covers the eastern half of the lower 48 states. We have obtained AWS wind power profile for 2006, it includes 1300 sites around NY Main River Area with both the hourly observed wind power and the hourly-predicted wind power for next 4 hours.

A sample of the wind power profile from one site is here

We simplify the GDC downtime model by setting up a threshold to the observed wind power without considering battery backup, server and services bootup time, etc. Different threshold will affect the total GDC downtime over a year and also the GDC downtime periods. Detailed analyzed data is shown in Tables below.

The table above indicates setting up the threshold to 0.1% of maximum wind power output over a year, GDC can gain a substantial high uptime to 98% of a year. Apparently when the threshold is set up 10% of maximum wind power over a year, GDC still can remain uptime to 83% of a year.

GDC downtime threshold is a critical parameter deciding GDC's availability and sustainable power usage efficiency.

From this table above, it says when threshold is at low bar, GDC will be more suffering frequent short downtime periods. However, when threshold is set at a high bar to 10%, GDC will experience the downtime by the daily weather conditions plus the short downtime periods.

Depending on the threshold, GDC may come up with different strategies for different downtime patterns. Accurate wind forecast is one of our mechanical group's focus in GDC project.